V: consonantal 'u' — new evidence

Latin Learning Latin
1

I’d like to add a bit to the debate regarding the pronunciation of consonantal ‘u’ in Latin, which is frequently repræsented by the letter ‘v’, such as in the name Vergilius or in the noun vīnum.

At the end of last year, I had travelled to the birthplace of my grandfather, a small hilltown called Guardiagrele, in the region the Romans called Aprutium, the Italians Abruzzo. A well known town in this mountanous, beautiful area is ancient Sulm?, the hometown of Ovid, which today we call Sulmona. I met some really wonderful people in Guardiagrele, but even there I was able to learn a lot about language. The local dialect, indeed, the language guardiese often has a lot more in common with Latin than modern Italian. For example, case is cheese, close to Latin caseus, while Italians call it formaggio properly. In any case, the people of the whole region in general, and especially in this very old town, have a few very distinct patterns of speech that are indicative of their ancient connexions with the Latin language.

Guardiagrele first came to be associated with the Romans by the Third Samnite War, during which the Romans successfully routed the Samnites who occupied the region, through force of arms and also by allying themselves with certain cities and tribes against the greater Samnite league. One of the towns that came to ally itself with Rome was the ancient antecedent to Guardiagrele, a union that brought great influence from Latium, including the language.

Enough of history. My observation in Abruzzo was, with great frequency, and particularly among the “less educated,” that everywhere a speaker would pronounce the letter ‘v’, instead he would enunciate a light ‘w’, as in English. Standard Italian ‘v’ is identical to that of English. Yet I would clearly hear the pronunciation doue for dove (“where”), moue for muove (“moves”), seruīr for servīre (“to serve”). I found this to be an astounding discovery. Clearly the ancient locution has been præserved, if only in more isolated populations.

Yet there is evidence that this antique version of the consonant is much more widespread than I first realized. I have taken note more recently from some Neapolitans the tendency to pronounce ‘v’ as ‘w’ when I speak with them in Italian (or in Neapolitan, the very little that I know). The dialects of Abruzzo share a good deal in common with other southern Italian dialects, and the permission of sharp Italian ‘v’ to flow into an older, gentler ‘w’ sound appears to be one of them.

It seems extremely convincing, therefore, that Classical Latin, or Latin of any period, may have pronounced consonantal ‘u’ as an English ‘w’.

2

that’s neat, Luce, I had always wondered why noone seemed to pronounce 'v’s as 'u’s anymore…
[I had written more, but upon further research found it to be wrong, so it is now gone]

-Jon

3

That’s a fascinating anecdote, Lucius. I wonder if there is anything in the literature discussing this tendency of rural Italians to speak a v that approximates the English w? It really does seem to have some bearing on the thorny u/v debate.

-David

4

An interesting observation - the level of diversity preserved in local Italian dialects is, it seems, remarkable and fascinating. But can we really use this as evidence for the pronunciation of Latin ‘v’?

You say that this village’s language has “more in common with Latin than modern Italian”. I’m reminded of suggestions that some small Appalachian communities are more like Shakespeare’s English than modern English. This seems to be the case because these communities preserve some features of Elizabethan English which standard English has lost – but the reality is, modern English preserves some features of Elizabethan English that these communities have lost. Any isolated group of language-speakers will preserve some features of a language while others change; compare two groups and each will have some features that, from the perspective of the other group, some more archaic.

You mention that this dialect preserves a form of the Latin ‘caseus’, which modern Italian does not; well, so does Spanish - does that mean Spanish is closer to Latin than Italian? The Romance languages are an excellent example of this principle; each preserves some features of Latin that the others don’t, and each has new developments moving away from Latin that the others don’t. In the examples you give, you mention that this dialect says ‘seruir’ where Italian says ‘servire’ - even if this dialect has preserved the ‘w’ sound of ‘v’, it has not preserved the final syllable of the infinitive, where Italian has.

Perhaps this dialect does preserve the original Latin pronunciation of ‘v’ - but it is also entirely possible that, even in this community, the sound changed to what English would call a ‘V’ and then back to ‘W’ again (we see ‘V’ turn to ‘W’, for example, in lower-class 19th century British English - it’s one of the normal stereotypes of uneducated pronunciation in Dickens). If one wanted to argue for ancient pronunciation like our ‘V’, one could imagine any number of scenarios under which it’s this dialect, not the other Romance languages, that has changed.

Don’t get me wrong - I do believe that Virgil pronounced his name “Wergilius” and not “Verjilius,” and I can appreciate the fascinating glimpse isolated dialects can give us into what normal Romans living 2000 years ago might have sounded like. But I don’t think you’re going to convince anyone who’s not already convinced with this one…

5

I may not have been clear about my intentions for this evidence, mraig. It was hardly to prove anything, merely to support the the possibility of Classical ‘w’-pronunciation. And the ancillary references to vocabulary left in common between Latin and Guardiese are purely circumstantial, and nothing more. This isn’t about vocabulary at all, or about the dropping of final vowels; it’s just the tendency of one little consonant — a behavior totally separate from other features.

6

I thought you might like this on Latin ‘u’.

A German writer, Johann Bodensteiner, wrote a book on speaking English in 1585 called Clavis linguae Anglicanae, è tabulâ in praesentem formam redacta. Certainly, it was for an educated European audience, or for those who spoke Latin and wanted to learn English.

Here’s what he says about pronouncing the letter W in English in his chapter on pronouncing letters, “De Literis”:
“W sicut u Latin. (in quando) VVade, vado: vve, nos: svvan, cygnus.” (p.9)
Nor does he give a ‘v’ sound distinct from a ‘u’ sound in his list of letters in English.
In other words, to this writer at least, a Latin ‘u’ = an English ‘u’ = an English ‘v’, and sounds exactly like an English ‘w’ at the end of the sixteenth century.

Note that ‘w’ was often written with ‘vv’ in the period, i.e., double ‘u’. Even in English, the ‘v’ = ‘u’ sound is why e’en = even, e’er = ever, ne’er = never and o’er = over, as the OED says. Because ‘v’ was also pronounced as ‘w’ in English in bygone times.
Adrianus

7

That’s amazing, Adriane! Then why did some 'w’s change into ‘v’ and others not?

This also means that, indeed, the pronunciation of even late Renaissance Latin clearly possessed the ‘w’ for the consonantal-u.

I’m sure you’re familiar with frequent Latin contractions: “norat” for “nouerat,” “audiit” for “audiuit,” and so forth.

8

Well, all you can say was that in some parts of Europe the ‘u’ sound was kicking around, and in other parts was a ‘v’ sound because of socio-linguistic diversities. There was the attempt to reconstruct and revive the “ancient Roman accent” by late-fifteenth-century humanists such as Erasmus. They felt it was needed, so clearly typical spoken later Latin was not along the classical model (however imagined) in many regards, but it may have been with regard to ‘v’ = ‘u’. All that Bodensteiner illustrates is one example of how a ‘u’ = ‘v’ sound was considered by some as a standard, --he takes it for granted. Nor did all English dialects pronounce ‘u’ = ‘v’, but it certainly wasn’t unusual and was, in fact, quite common,–even though it finally became something of an upper-class affectation, I think, to pronounce your ‘v’ as a ‘u’. An historical linguist can probably tell you right away; or a specialist in Elizabethan literature and pronunciation. I’ve got a lot of articles on it but I haven’t been reading them from that point of view–just for clues to syllable accenting, accent notation and punctuation.
Adrianus
P.S. Yes, the contractions are interesting for me because, in their contracted forms, the accenting in certain cases --‘audíit’, say–contradicts the penultimate rule (as acknowledged by the grammarians–well, it does and it doesn’t, because the accented ‘i’ started off as a long ‘i’, but it shouldn’t be a long ‘i’ before another ‘i’ in ‘ii’, so gets shortened but its still accented!!). And you’re right, they parallel the English syncopated examples very nicely.

P.P.S. I looked back at one of the English grammar reformers Bullokar, his Booke at Large for the Amendment of Orthographie for English Speech (London, 1580) and he says nothing about consonantal-u being pronounced as w in English, just as v. So the v = w sound was long gone, except in some dialects. Bullokar also says (p.7) that, when speaking Latin, a Latin consonantal-u is sounded in England with a ‘v’ sound–so he’s against the reformers. Lloyd’s Encyclopaedic Dictionary of 1895 notes v = u (or w) in ‘periwinkle’ from French ‘pervenche’ Latin ‘perivinca’ and in “vulgar speech, especially of Londoners”, as Mraig says above about Dickensian London. It isn’t impossible that Bodensteiner can’t pronounce a ‘v’ and so teaches all his readers the latin ‘u’ for speaking English consonantal ‘v’.
P.P.P.S. On closer reading, I think Bodensteiner has just forgotten to discuss the ‘v’ sound. Apologies, Luce. I got carried away.

9

I cannot understand something:

if the letter “v” was standing for the vowel in the latin words as “venio, venire”,
why the generated words are pronounced with the consonant [v] in all the romance languages, from Portuguese to Romanian?
the words “vir” (pt), “venir” (es), “venir” (fr), “venire” (it), “veni” (ro) are written with “v” and pronounced with the consonant [v] in all these languages.

maybe those words were pronounced with [v] in vulgar latin?

thx

10

Actually, it stood for [w], a semi-vowel.

why the generated words are pronounced with the consonant [v] in all the romance languages, from Portuguese to Romanian?
the words “vir” (pt), “venir” (es), “venir” (fr), “venire” (it), “veni” (ro) are written with “v” and pronounced with the consonant [v] in all these languages.

maybe those words were pronounced with [v] in vulgar latin?

Sound change, of course. Originally V was pronounced [w], which was inherited from Proto-Indo-European. For example, “vir” and “vesper” are congate with English “werewolf” and “west”. In Vulgar Latin, V would have become [β] (a voiced bilabial fricative) before becoming [v].

11

yes, I meant [w], shame on me I wrote .

In Vulgar Latin, V would have become [β] (a > voiced bilabial fricative> ) before becoming [v].

thank you very much!

and thanx once more, because now I understand better something about the Spanish language.


in the meantime, I realized that some Latin words, written with “v”, generated Romanian words with [w] (written “u”), as “novem” → “nouă”.

12

I know I might be bumping an old thread, but I think we should consider Romanian as a good evidence.

bos bovis(ox) is bou in Romanian and is expressed more like English “bow”, but with an o umlaut.

I think Romanian is unfortunately too often neglected, when doing comparative linguistics/philology between Romance languages.

13

I don’t know why this is so important to me. I have to end this obsession… :blush: Anyway, I was looking into a letter that Pope Pius X wrote to an Archbishop recommending the use of ecclesiastical latin (which, I’m now understanding, was not italianate before the XXth century), and I came upon this excellent article:

http://avitus.alcuinus.net/schola_latina/soni_es.php

After reading about “el babel eclesiástico”, I went immediately looking for the semiconsonants ‘j’ and ‘v’, and, this is what it says (in the Spanish version):

La ‘v’ […] suena como la ‘w’ inglesa en wet (> más probablemente > como la व del hindi). Los hablantes de español en particular deben esforzarse por no pronunciarla como [gw]. Es probablemente mejor (ciertamente más justificado desde el punto de vista histórico) pronunciar ‘vís’ como ‘bis’ (con nuestra suave [β] > fricativa> ) que pronunciarla [gwɪs] por querer decir [wɪs] y no poder.

[The emphasis is mine.]

If you look up Hindi in Wikipedia (http://en.wikipedia.org/wiki/Hindi), you’ll find व to be an Approximant, between ‘w’ and ‘v’.

Valete! :slight_smile:

14

Indeed, Amadeus, and I’ve suggested this possibility myself not a few times. I wish there were some Indian-language experts around. As far as I understand, the ‘va’ of Hindi (pronounced “wa”) in indeed between sounds. However, it shows its range of character in different environments: ‘va’ + ‘e’ or ‘i’ = “ve,” “vi;” ‘va’ + ‘a’, ‘o’, or ‘u’ = “wa,” “wo,” “wu” — approximately; at least, that’s how it sounds to the English ear (the strong Hindi accent has these traits). The Romans never announced something so distinct. However, an Indian most of the time will not be aware of this distinction to our ears, and indeed will insist that the sound is neither v nor w, and unchanged in environments of different vowels. Perhaps the Romans were in the same boat.

Similarly, the consonant in the syllables ‘ki’, ‘ka’, and ‘ku’ sound the same to all our ears, when in reality they are all formed in different places in the mouth due to their respective vowel sounds. Although we don’t hear a distiction right away, the Etruscans did, and had three separate letters for each vowel: ci, ka, qu — hence the use even today of ‘q’ only before ‘u’; it was one of the Etruscan traits that the Romans maintained.